TransExample2011

TransExample2011 - # 19 ECE 253a Digital Image Processing...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: # 19 ECE 253a Digital Image Processing Pamela Cosman 11/17/11 Quantitative Example of Energy Compaction and Decorrelation Let u be a 2 × 1 vector with mean zero: u = bracketleftBigg u (0) u (1) bracketrightBigg and covariance R u = parenleftBigg E [( u (0)- μ )( u (0)- μ )] E [( u (0)- μ )( u (1)- μ 1 )] E [( u (0)- μ )( u (1)- μ 1 )] E [( u (1)- μ 1 )( u (1)- μ 1 )] parenrightBigg = parenleftBigg E [ u (0) 2 ] E [ u (0) u (1)] E [ u (1) u (0)] E [ u (1) 2 ] parenrightBigg = parenleftBigg 1 ρ ρ 1 parenrightBigg for 0 < ρ < 1. From the expression for R u , the variances σ 2 u (0) = σ 2 u (1) = 1 that is, the total average energy of 2 is distributed equally between u (0) and u (1). The parameter ρ gives an indication of the correlation between u (0) and u (1). The correlation is by definition corr [ u (0) , u (1)] = E [ u (0) u (1)] σ u (0) σ u (1) = ρ So in this case, the off-diagonal elements of the covariance matrix are in fact exactly equal to the correlation, but that is only because the...
View Full Document

This document was uploaded on 01/20/2012.

Page1 / 3

TransExample2011 - # 19 ECE 253a Digital Image Processing...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online